National Oil Companies: Strategy, Performance and Implications for Global Energy Markets Prospectus

Program on Energy & Sustainable Development Stanford University August 15th 2006

While the role of the state is declining in nearly every sector of world economic activity, in hydrocarbons the pattern is quite different. State-controlled oil companiesso-called national oil companies (NOCs)remain firmly in control over the vast majority of the worlds hydrocarbon resources. In some countries where state companies had been waning in influence as governments adopted rules to encourage investment by private firmsincluding the international oil companies (IOCs)recent years have seen a resurgence in the roles for statecontrolled companies. High energy prices have encouraged governments and NOC managers, alike, to concentrate authority in the hands of state firmsespecially in Russia and Venezuela while many countries have shelved plans to open their hydrocarbon sectors. The central role for NOCs seems more secure than ever. Governments are struggling to find the right policies toward their NOCs. The IOCs are struggling to find their role in this setting. This study will explore the factors that explain the organization and performance of the worlds largest NOCs. It will involve in-depth case studies of about 15 NOCs that include the enterprises that control the worlds largest oil reserves as well as several NOCs that have large refining and marketing operations. This prospectus summarizes the goals and methods of the study; a sample outline for a typical case study is included in an appendix. NOCs are hardly monolithic. Some of these enterprises are singular in their control over their home market while others are exposed to competition in some or all the segments of the value chain from exploration & production of hydrocarbons to refining and marketing. Some engage in joint ventures while others create their needed capabilities in-house. Some operate internationally while others stay at home. Some operate projects while others engage passively. These enterprises differ markedly in the ways they are governed and the tightness of their relationship with government. NOCs also vary in their geological gifts, as some are endowed with prodigious quantities of easy oil while others must work harder and apply highly advanced technologies. Some have sought natural gas, which requires different skills and market orientation than oil, while others stay focused on liquids. This study will explore how such factors affect the performance of NOCs. Our aim in this work is partly social science, partly a study of business strategy, and partly normative. As social scientists we note that NOCs are organized and behave in ways that 1

are strikingly different from private (non-state) firms that are the basis of most theories of industrial organization. NOCs are often, in effect, states within a state. As large suppliers of state revenues they have a special role in state budgets; usually they are among the most attractive employers in the country; typically they exert influence on a wide range of energy services (e.g., electric power supply) in addition to hydrocarbons. These special roles confer enormous political power that NOCs have used in highly varied ways. Our goal is to improve the theory that explains how such special enterprises interact with their host governments. Concerning business strategy, we are interested in how NOCs might engage most productively with other enterprisesincluding, notably, the IOCs. Many IOCs have made bold claims about their special role as suppliers of technology, capital and access to markets that are difficult to square with reality, especially in the present high price environment where many governments have not allowed IOCs access to hydrocarbon resources. NOCs have often found it difficult to craft viable partnerships with outside firms; most NOCs have found it especially difficult to operate outside their home market. We will explore which business strategies have been attempted and which prove viable. Concerning policy, we want to understand how governments can influence the behavior of NOCs. There is a long history of governments reorganizing their hydrocarbon sectors with the goal of improving governance. Western oil companies were originally invited into oil-rich parts of the world in part because host governments thought it would be easier to control development of the sector by working with outsiders that had expertise and capital. The shift to NOCs, which occurred mainly from the 1970s through the early 1980s, was pursued partly on the same logic. Yet both modes of organization have their own troubles. Governments have incomplete information about the real cost and opportunity for developing hydrocarbon resources and they have incomplete leverage over inside and outside firms alike. Governments are taking a fresh look at best practices, and many are encouraging a hybrid of NOCs and private firms. Our study will document the varied outcomes from these different policy efforts and suggest which approaches have the best prospects for governments in the future. Indeed, some countries have imposed reforms on their NOCs that have been extremely successful in boosting output and improvement the effectiveness with which these firms employ capital, technology and labor. Effective policies are important because NOCs control about three-fourths of the worlds proven oil reserves and there is little indication that NOCs will be on the wane.

What we want to explain We aim to explain the varied experiences of NOCs along two dimensions. First, and most important, we want to explain performance, by which we mean the economic efficiency with which NOCs find, develop and deliver hydrocarbon resources. We will use conventional measures of hydrocarbon performance such as cost and employment per barrel found, produced and refined. We will focus on measures of economic performance and will report (or estimate) orthodox financial measures of performance (e.g., return on capital employed) and will also adjust those measures for the varied social obligations and differences in taxation that vary across firmsso that we can examine and compare the NOCs on equivalent basis. We are mindful that most NOCs are tight-lipped about the underlying data that would be needed to 2

assess performance. Thus our studies will combine standard indicators (where available) such as cost per barrel found and produced as well as employment per output. We will augment those data with subjective assessments from a wide range of expert observers obtained through structured interviews. Second, we aim to explain NOC strategy at home and, where relevant, in markets abroad. What kinds of partnerships have the NOCs entertained and embraced? We will focus, especially, on equity partnerships, operational partnerships, and field service partnerships. To obtain information on these partnerships we will rely heavily on field interviews and work done in the NOC countries and companies. In evaluating these partnerships we will imagine how an efficient firm would deploy partnerships as part of a broader strategy to manage risk and to concentrate in areas of comparative advantage. Such a firm wouldthrough investments, acquisitions and divestituresdevelop a portfolio of projects to extract and process hydrocarbon resources, with partnerships to complement core capabilities. Such a portfolio would help it manage market, political and technological risks. We will compare the actual pattern of the NOCs partnerships with such a hypothetical efficient benchmark.

Main Factors In each case we expect that a wide range of factors will affect performance and partnership strategies in complex ways. At the outset of this study we did not know which factors would be most important, so we identified four categories of factors that probably explain NOC performance and partnerships. These factors are often cited in the literature on NOCs (and on state firms less specifically), and thus our study offers an opportunity to test them with systematic data from a large number of cases. Each of the case studies will document the first three of these categories of factors, along with a history of the NOC and an analysis of the fuller range of factors at work. In addition, the project team will gather information about the fourth factor, which concerns the structure of the state, and use it at the end of the study to examine systematically how such factors may influence outcomes.

1. State goals, capabilities and relationship with the oil sector. Governments created NOCs, mainly from the 1950s, with particular goals in mind. Some initially sought to create a benchmark that could be used to augment their efforts to regulate and tax foreign and private firms operating in the country. Others went directly to the full endpointa national firm with exclusive controlbecause they thought the state could obtain maximum revenue and leverage only if it monopolized all aspects of production. Such decisions arose in the context of a shift toward greater state control in most economies, with the lucrative oil sector considered of paramount importance. We seek information on three major aspects of the relationship between the state and the sector. First, what are the goals of the state for the sector and what strategies has government followed to pursue those goals? Each study will examine this question over time, with attention

to factorssuch as the price of oil (and thus level of rents)that may explain how those goals have changed. Second, we examine the particular instruments that government has deployed to affect behavior in the sector, with attention to regulation and competition policy in particular. Regulation includes price controls as well as health, safety and environmental (HSE) rules. Competition policies affect the number (if any) of competing firms able to enter the sector and the terms of competition. Together, these factors affect the industrial structure and, plausibly, will have a large impact on performance and strategy. We focus on the capabilities of government to develop and apply regulatory and competition policy as performing these functions requires obtaining extensive information needed to monitor the behavior of the enterprise and to make decisions. Where the firms are formally part of government, performing such functions can be especially difficult because government has a conflict of interest and information may be especially concentrated inside the NOC itself. Yet, to allow the firm to behave efficientlyand for a market to function in the countries that have sought to encourage market competitionthese capabilities must be developed and applied in firm, predictable and often subtle ways. Third, we examine the fiscal relationships between the state and the NOC. Each study will summarize the main taxes (and other fees) paid by the NOC and subsidies. Each will also examine the fiscal controls, if any, applied by the state to capital spending by the NOC. Such factors will help to explain the importance of the NOC to the financial health of the state (and the magnitude of resources at stake, which is likely to influence efforts to control the NOC). Taxation, as well as controls on capital spending, will help to explain the investment patterns of the NOC and thus its performance and strategy.

2. Management Next, we examine a series of factors related to the governance and control of the NOC itself. We examine aspects of management. First, we examine the formal procedures for governance, with particular attention to several points of control: Formal ownership and protections for minority shareholders (if any); Control over executive and other senior appointments; Formal and de facto decision-making, with particular attention to control over the process for proposing and approving new investments. Disclosures and transparency, such as standards for financial accounting (e.g., SEC rules for listed companies; alternative rules for non-listed enterprises)

We think these factors matter because they ultimately affect the identity of key decision-makers, the masters they serve, the information that the enterprise supplies to government and markets, 4

and critical decisions such as investments. In turn, that should affect firm performance and choice of partnerships. For example, a strong staff with robust internal procedures can keep the firm focused on commercially viable projects, making the enterprise more immune to external political meddling. Second, the studies will examine each NOCs sources of investment capital. Is the NOC able to raise capital on world markets (what is its bond rating, can it earmark overseas capital against hydrocarbon revenues, etc.), can it retain earnings, and how (if at all) does it draw its capital budget from the state (and at what cost)? These factors will affect the ability of the NOC to mobilize and control capital for investment. Third, we examine the organizational structure and culture of the NOC. We examine the formal organization of the firm and actual lines of authority and promotion within the firm. Each case study will describe training and typical career paths of key officials and the esprit de corps within the enterprise (which, in turn, relates to whether/how key staff positions are awarded by merit or patronage). Where possible, the studies will explain whether some parts of the NOCs organization have stronger esprit de corps and lead to better performance than others, and why.

3. Technology and Hydrocarbon Resources. Each study will also examine the interplay of technology and the cost/availability of the hydrocarbon resources that the NOC is able to tap. Where the NOCs have R&D programs of their own, each case study will examine where those R&D programs have succeeded and failed. We also ask where the enterprise obtains its hydrocarbon resources and the size of those resources. Does it have exclusive access to hydrocarbon resources within its national territory, does it face competition? Must it pay for its resources within its national territory, or are they provided at little or no cost? In acquiring acreage overseas, does the firm rely on special political deals and access afforded through its connections to the government, or does it compete on an even keel with other firms?1 We expect that provisions for access to resources will affect the firms E&P strategy both domestically and overseas; moreover, it should affect firm performance since competition may yield improvements in economic efficiency. Control over resources in the ground may also be an instrument that government can use to exert leverage in the sectorto affect the behavior of the NOC and (where relevant) other competitors.

We should be attentive to situations where the firm faces different resource rules at home and abroad as that, in turn, may yield different levels of efficiency in its domestic and overseas E&P operations. (Or it may yield the need for internal subsidies or some such arrangement if, for example, the firm must compete and pay for acreage overseas but is unable to operate such fields efficiently.)

We are mindful that these studies must be attentive to the differences in geology and the nature and type of resources under the control of state companies. Is the resource easily accessible, or is advanced technology required for extraction? What is the quality of the resource? (To help answer these questions systematically we are developing a database of published information on these countries production costs and international benchmarks.) We also examine the differences between oil and gasat least in cases where the national company has both types of resources. The market for oil is nearly guaranteed for all firms that successfully develop their resources. Gas is quite different as merely having the resource doesnt mean it will get used. The market for gas depends on the vagaries of other competing fuels (by contrast, oil has essentially no competitors) and technologies. Increasingly, gas is being used for electric power; thus, assuring a market for gas requires managing regulatory relationships. Attention to the issues surrounding management of technology and its interplay with hydrocarbon resources will allow us to address questions such as: how do decisions by government affect which resources are developed and at what cost? to what degree is the NOCs performance explained by the resources that were tapped for example, did the NOC develop advanced technology (and an engineering culture) because it was required to tap difficult resources? has the NOC followed different strategieswith different roles for and types of partnerships and internal development of technologyin tapping different types of resources?

4. State Structure Most of this study is focused on the attributes of the NOC, the resources it has available to tap, and the relationships between the NOC and government. These factors, we think, will explain the variation in performance of the NOC. However, we must be attentive to the possibility that the root causes of NOC performance are found elsewhereperhaps in the structure of the state and its system of government. Thus, in parallel with the development of the 15 case studies, a side project will examine the major factors related to state structure and will test whether they coincide with the performance and strategy of the NOCs. This part of the study awaits information on performance and strategy from the 15 case studies and will also require a systematic review of the literature, but the factors we are likely to examine include: Percent of GDP, budget and exports accounted for as hydrocarbons; Legislative control over law-making and budgeting; Deficit or surplus of state budget; 6

Standard variables of governance, such as transparency and corruption;

We remain skeptical that such factors are powerful explanators, but these hypotheses need to be tested.

Units of Analysis and Case Selection The empirical research in this study will use NOCs as the unit of analysis. By NOC we mean a state-owned enterprise engaged in the production or marketing of oil or gas within a national territory. By state-owned we mean an enterprise (firm) for which the government has a controlling interest100% in some cases, but in a few cases as low as perhaps 20% if other shares are held widely or if the state has concentrated ownership of voting shares. For some firms, the state has a golden share or other arrangement that confers state control and thus qualifies as a state-owned enterprise in our definition. To select our sample we compiled a long list of all significant state-owned oil companies (appendix B).2 From that list we then selected a sample of 15 NOCs that span the range of the worlds experience with the four categories of factors identified above. These factors are difficult to measure systematically, and our assessments of the variation were subjective. Thus we have erred on the side of selecting companies that are intrinsically important so that our sample was not only scientifically valid but also obviously relevant for the largest companies and host countries. After compiling the full information about the range of experiences we will recheck and possibly adjust our case selection. 3 Table 1 shows our sample of 15 NOCs.

Work Plan The study will run from spring 2006 and run for 18 months, with the effort in stages. This prospectus is being written in the middle of the first stage where we are testing the research protocol with initial studies on six NOCs that have quite different functions and histories (PdVSA, Pemex, Petrobras, NIOC, CNPC and Gazprom). Based on those results we will revise the protocol late in 2006 and begin work on the full sample of companies. We plan a large review meeting in late April 2007 to discuss all the major factors identified in this prospectus, to engage industry and academic observers, and to present initial drafts of the six trial case studies.

2 3

This list is drawn from Petroleum Intelligence Weekly [complete citation].

A similar process (without the final readjustment phase) was followed in PESDs study on independent power producers (Woodhouse, A Political Economy of International Infrastructure Contracting: Lessons from the IPP Experience, PESD Working Paper #52, 2005).

The final stages of work will include analysis of the factors related to state structure and will also synthesize the findings and then apply them to some critical questions in the industry, such as the ways to manage joint ventures with NOCs, and possible responses from the NOCs to a decline in oil prices.

Table 1
Gas Liquids Output Output (MMcf per (1000 bpd) day) 9,045 3,852 2,170 2,500 1,200 2,166 3,723 221 1,729 2,120 1,701 731 557 332 740 6,900 7,640 1,054 4,000 4,242 677 3,244 52,244 7,807 2,407 2,010 4,172 2,486 6 1,921

Enterprise Name Saudi Aramco NIOC (National Iranian Oil Co) KPC (Kuwait Petroleum Co) PdVSA (Petroleos de Venezuela) ADNOC (Abu Dhabi National Oil Company) NNPC (Nigerian National Petroleum Co) Pemex Gazprom Sonatrach CNPC* Petrobras Petronas ONGC Sonangol Statoil

Country Saudi Arabia Iran Kuwait Venezuela U.A.E. Nigeria Mexico Russia Algeria China Brazil Malaysia India Angola Norway

Oil & Gas Reserves (Bboe) 300.9 295.2 109.0 104.4 79.2 40.2 18.7 191.5 37.3 18.4 11.1 24.9 6.6 2.2 4.3

Source: Petroleum Intelligence Weekly, 2003 (table to be updated with new PIW survey) *CNPC reserves and output reported as PetroChina

Annex A: Universe of Significant NOCs

Enterprise Name Saudi Aramco NIOC (National Iranian Oil Co) INOC (Iraqi National Oil Council) KPC (Kuwait Petroleum Co) PdVSA (Petroleos de Venezuela) ADNOC (Abu Dhabi National Oil Company) Libya NOC NNPC (Nigerian National Petroleum Co) Pemex Qatar Petroleum (QP) Gazprom PetroChina Sonatrach Petrobras Petronas Pertamina Petroecuador ONGC Sinopec PDO Socar Rosneft Kazmunaigas Sonangol Syrian Petroleum (SPC) EGPC Statoil Ecopetrol CNOOC Norsk Hydro PetroKazakhstan PTT Forest Oil INA CPC IOC (IndianOil)

State Ownership (%) 100% state 100% state 100% state 100% state 100% state 100% state 100% state 100% state 100% state 100% state 38.37% state 90% state 100% state 51% state 100% state 100% state 100% state 84% state 55% state 60% state 100% state 100% state 100% state 100% state 100% state 100% state 82% state 100% state 71% state 43.8% state private 100% state 78% state 100% state 100% state 82.03% state

Country Saudi Arabia Iran Iraq Kuwait Venezuela U.A.E. Libya Nigeria Mexico Qatar Russia China Algeria Brazil Malaysia Indonesia Ecuador India China Oman Azerbaijan Russia Kazakstan Angola Syria Egypt Norway Colombia China Norway Kazakstan Thailand Finland Croatia Taiwan India

Oil & Gas Reserves (Bboe) 300.9 295.2 134.8 109.0 104.4 79.2 44.3 40.2 18.7 133.143 191.5 18.4 37.3 11.1 24.9 20.96916 4.5621 6.64662 3.77684 6.5068 5.535 27.1806 2.514 2.232 2.8958 7.39152 4.28848 2.2692 2.18372 2.31006 0.35368 1.03666 0.22644 0.1995 downstream only downstream only

Gas Liquids Output Output (MMcf per (1000 bpd) day) 9,045 6,900 3,852 7,640 1,330 239 2,170 1,054 2,500 4,000 1,200 4242 896 617 2166 677 3,723 3,244 660 2,144 221 52,244 2120 2,407 1729 7,807 1701 2010 731 4,172 1139 2562 204 10 557 2486 742 514 702 1284 178 500 393 678 150 130 332 6 317 510 375 1611 740 1921 292 594 306 291 395 753 151 19 555 24 266 24 179

10